CN112507970A - Unmanned car washing image recognition anti-collision system and recognition method based on PointRend algorithm - Google Patents
Unmanned car washing image recognition anti-collision system and recognition method based on PointRend algorithm Download PDFInfo
- Publication number
- CN112507970A CN112507970A CN202011553858.2A CN202011553858A CN112507970A CN 112507970 A CN112507970 A CN 112507970A CN 202011553858 A CN202011553858 A CN 202011553858A CN 112507970 A CN112507970 A CN 112507970A
- Authority
- CN
- China
- Prior art keywords
- algorithm
- vehicle
- collision
- washed
- image recognition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005406 washing Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000004891 communication Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000002618 waking effect Effects 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 description 4
- 238000012790 confirmation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
- G06V20/586—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of parking space
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60S—SERVICING, CLEANING, REPAIRING, SUPPORTING, LIFTING, OR MANOEUVRING OF VEHICLES, NOT OTHERWISE PROVIDED FOR
- B60S3/00—Vehicle cleaning apparatus not integral with vehicles
- B60S3/04—Vehicle cleaning apparatus not integral with vehicles for exteriors of land vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30264—Parking
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of unmanned vehicle washing image identification, and particularly relates to an unmanned vehicle washing image identification anti-collision system and an unmanned vehicle washing image identification method based on a PointRend algorithm. The system comprises a vehicle to be washed, a car washer, an image acquisition device, an image recognition anti-collision algorithm system and an information feedback system; the image acquisition device, the image recognition anti-collision algorithm system and the information feedback system are sequentially in communication connection; the image recognition anti-collision algorithm system comprises a PointRend algorithm module, the body outline of the vehicle to be washed is recognized through the PointRend algorithm, whether the vehicle to be washed is out of range or not is judged in advance through presetting an anti-collision reference line in the image to be crossed with the body outline, and a feedback result is transmitted to the information feedback system. The invention has the characteristics of accurately judging the parking position of the vehicle in the unmanned vehicle washing device, timely informing a user of righting the vehicle and preventing the vehicle from being collided.
Description
Technical Field
The invention belongs to the technical field of unmanned vehicle washing image identification, and particularly relates to an unmanned vehicle washing image identification anti-collision system and an unmanned vehicle washing image identification method based on a PointRend algorithm.
Background
An unmanned car washer refers to a car washer which does not need car washing personnel to participate in car washing in the whole process, and has the biggest characteristic that intelligent equipment is used for replacing traditional car washing manpower, and the unmanned car washer has the characteristics of high car washing efficiency, low car washing cost and the like. The car washing using process of the conventional unmanned car washer on the market at present is as follows: after the vehicle to be washed is paid autonomously, the vehicle enters a parking area of the unmanned vehicle washer, and vehicle washing equipment in the unmanned vehicle washer is triggered to set a vehicle washing rule to wash the vehicle to be washed.
However, the prior unmanned car washer has the following technical problems in the actual operation process: in the process that a vehicle to be washed automatically runs into the unmanned car washer, the vehicle may accidentally collide with the unmanned car washer due to the problem of the parking position of the vehicle, and further the unmanned car washer is damaged. The car washing equipment in the unmanned car washer is intelligent controlled precision equipment which is easily damaged unrecoverable once being collided by a car, so that the operation cost of the unmanned car washer is increased; meanwhile, after collision occurs, the vehicle is inevitably damaged and needs to be maintained, so that unnecessary troubles are brought to the vehicle owner.
For example, the multiple positioning car washing system and the car washing method described in chinese patent application No. CN201811600774.2 include a car washing room, a car information recognition device, a positioning device, a big data processing center, a car washing device, a water circulation device, and a control device; the positioning device comprises a first positioning device and a second positioning device. When a vehicle to be washed enters a garage for washing, the vehicle information recognition device recognizes the license plate number, and the system automatically calls corresponding vehicle type data; the first positioning device guides the vehicle to park; the second positioning device scans the appearance of the vehicle and transmits information to the big data processing center for comparison, and if the information is inconsistent, the information is modified and stored again according to the scanning result of the second positioning device, so that the vehicle type can be attached more accurately during washing. Although the vehicle information recognition device, the first positioning device and the second positioning device are matched with each other, so that the system can perform multiple positioning confirmation on the vehicle parking position and the vehicle type information, and the vehicle washing efficiency and the cleanliness are improved, the system has the defects that the system only performs multiple positioning confirmation on the vehicle parking position and the vehicle type information, and does not consider the situation that the vehicle is easy to collide when the vehicle parking position is incorrect, and a corresponding early warning system is lacked.
Disclosure of Invention
The invention provides an unmanned car washing image recognition anti-collision system and a recognition method based on a PointRend algorithm, which can accurately judge the parking position of a vehicle in an unmanned car washing device, can timely inform a user of righting the vehicle and prevent the vehicle from being collided, and aims to solve the problems that in the prior art, the process of the vehicle to be washed entering a parking area by self-running and the position of the vehicle parked are difficult to control under the condition that the vehicle to be washed is not guided in the actual running process of the existing unmanned car washing machine, and the vehicle is easy to collide with the unmanned washing machine.
In order to achieve the purpose, the invention adopts the following technical scheme:
the unmanned car washing image recognition anti-collision system based on the PointRend algorithm comprises a car to be washed, a car washer, an image acquisition device, an image recognition anti-collision algorithm system and an information feedback system; the image acquisition devices are arranged at the two sides in front of or behind the car washing station and are used for acquiring photos of the two sides of the car body to be washed; the image acquisition device, the image recognition anti-collision algorithm system and the information feedback system are sequentially in communication connection; the image recognition anti-collision algorithm system comprises a PointRend algorithm module, identifies the vehicle body outline of the vehicle to be washed through the PointRend algorithm, pre-judges whether the vehicle to be washed crosses the boundary or not by presetting whether an anti-collision reference line is crossed with the vehicle body outline or not in the image, and transmits a feedback result to the information feedback system; the information feedback system is used for informing a feedback result to a user.
Preferably, the information feedback system comprises at least two pairs of correlation electro-optical sensors, a Programmable Logic Controller (PLC), an Internet of things module, an upper computer, an LED display screen and a server; the server is connected with the Internet of things module, the Internet of things module is connected with the PLC, the PLC is connected with the image acquisition device and each correlation photoelectric sensor respectively, the upper computer is connected with the Internet of things module, and the LED display screen is connected with the PLC.
Preferably, the correlation electro-optical sensor and the PLC are both arranged on a car washer.
Preferably, the image acquisition device comprises at least two cameras; the cameras are arranged on the two sides in front of or behind the car washing station; the cameras are in communication connection with the image recognition anti-collision algorithm system.
The invention also provides an identification method of the unmanned car washing image identification anti-collision system based on the PointRend algorithm, which comprises the following steps:
s1, driving the vehicle to be washed into the station by the user, and waking up the image recognition anti-collision algorithm system by the user through a code scanning wake-up algorithm or a PLC state read wake-up algorithm of the car washer;
s2, after the algorithm is awakened, the unmanned vehicle washing image recognition anti-collision system collects all pictures shot by the cameras and submits the pictures to an interface of the image recognition anti-collision algorithm system;
s3, setting coordinates of two pixel points as coordinates of two ends of a reference line through parameter setting by the image recognition anti-collision algorithm system in a picture shot by a camera, calculating the outline of the vehicle to be washed by using a PointRend algorithm, and comparing whether the outline of the vehicle to be washed is crossed with the reference line;
and S4, the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 and informs the user.
Preferably, the code scanning wake-up algorithm in step S1 includes the following steps:
when a user scans a code, the server acquires a code scanning action of the user, reads the PLC state of the car washer through the Internet of things module, controls the camera to capture an image if the PLC acquires a parking signal of the correlation photoelectric sensor, and wakes up an algorithm to start calculation; if the PLC does not obtain the parking signal of the opposite-emitting photoelectric sensor, the user is prompted through an LED display screen or a mobile phone, the vehicle to be washed is parked in the specified front and rear limited parking areas until the PLC obtains the positions of the front and rear parking areas, and then the image recognition anti-collision algorithm system is awakened.
Preferably, the step S1 of reading the car washer PLC state wake-up algorithm includes the following steps:
the correlation photoelectric sensor is used for detecting whether the vehicle to be washed enters a front and rear limited parking area, and if a signal fed back to the PLC by the correlation photoelectric sensor indicates that the front and rear limited parking area is indicated, the front and rear position of the vehicle belongs to a safe area; and the upper computer reads the stop signal of the PLC through the Internet of things module and awakens the image recognition anti-collision algorithm system.
Preferably, the step S3 further includes the steps of:
drawing a straight line between the coordinates of the two pixel points set by the parameter setting to be used as a reference line;
if any pixel point is superposed with any coordinate point on the reference line in the outline of the vehicle to be washed drawn by the PointRend algorithm, the vehicle to be washed is considered to be out of range; and if all contour pixel points of the vehicle to be washed are not superposed with any coordinate point on the reference line, the vehicle is considered not to be out of range.
Preferably, the step S4 further includes the steps of:
the user wakes up the algorithm by scanning the code, and the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 to the mobile phone end of the user and prompts whether the vehicle to be washed is out of bounds; and (4) the user wakes up the algorithm by reading the state of the car washer PLC, the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 to the PLC, and the PLC outputs a comparison result signal to the LED display screen and informs the user whether the car to be washed is out of range or not.
Compared with the prior art, the invention has the beneficial effects that: (1) the system can accurately judge whether the parking position of the vehicle to be washed in the unmanned vehicle washing device is out of range or not, and timely inform a user of vehicle alignment so as to prevent the vehicle from being collided; (2) the invention applies the PointRend algorithm to an unmanned car washing image recognition system, recognizes the car body outline through the algorithm, pre-judges whether the car is out of range or not by presetting whether an anti-collision reference line and the car body outline are crossed or not in the image, and feeds back the result, thereby having the characteristics of high efficiency and high quality.
Drawings
FIG. 1 is a flow chart of an identification method of an unmanned car washing image identification anti-collision system based on a PointRend algorithm in the invention;
fig. 2 is an effect diagram of the identification method of the unmanned car washing image identification anti-collision system based on the PointRend algorithm.
In the figure: the vehicle contour 1 to be washed, reference line 2.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention, the following description will explain the embodiments of the present invention with reference to the accompanying drawings. It is obvious that the drawings in the following description are only some examples of the invention, and that for a person skilled in the art, other drawings and embodiments can be derived from them without inventive effort.
Example 1:
the unmanned car washing image recognition anti-collision system based on the PointRend algorithm comprises a car to be washed, a car washer, an image acquisition device, an image recognition anti-collision algorithm system and an information feedback system; the image acquisition devices are arranged at the two sides in front of or behind the car washing station and are used for acquiring photos of the two sides of the car body to be washed; the image acquisition device, the image recognition anti-collision algorithm system and the information feedback system are sequentially in communication connection; the image recognition anti-collision algorithm system comprises a PointRend algorithm module, identifies the vehicle body outline of the vehicle to be washed through the PointRend algorithm, pre-judges whether the vehicle to be washed crosses the boundary or not by presetting whether an anti-collision reference line is crossed with the vehicle body outline or not in the image, and transmits a feedback result to the information feedback system; the information feedback system is used for informing a feedback result to a user.
The PointRend algorithm module can be applied to the tasks of instance segmentation (such as Mask R-CNN) and semantic segmentation (such as FCNs). For example segmentation, PointRend is applied to each region, and the mask is computed from coarse to fine by predicting a set of selected points. For semantic segmentation, the PointRend algorithm module can treat the whole image as a region.
Further, the information feedback system comprises two pairs of correlation electro-optical sensors, a Programmable Logic Controller (PLC), an Internet of things module, an upper computer, an LED display screen and a server; the server is connected with the Internet of things module, the Internet of things module is connected with the PLC, the PLC is connected with the image acquisition device and each correlation photoelectric sensor respectively, the upper computer is connected with the Internet of things module, and the LED display screen is connected with the PLC. Wherein, the correlation electro-optical sensor and the PLC are both arranged on the car washer.
Further, the image acquisition device comprises at least two cameras; the cameras are arranged on the two sides in front of or behind the car washing station; the cameras are in communication connection with the image recognition anti-collision algorithm system.
Based on embodiment 1, the invention also provides an identification method of the unmanned car washing image identification anti-collision system based on the PointRend algorithm, as shown in fig. 1, the identification method comprises the following steps:
s1, driving the vehicle to be washed into the station by the user, and waking up the image recognition anti-collision algorithm system by the user through a code scanning wake-up algorithm or a PLC state read wake-up algorithm of the car washer;
s2, after the algorithm is awakened, the unmanned vehicle washing image recognition anti-collision system collects all pictures shot by the cameras and submits the pictures to an interface of the image recognition anti-collision algorithm system;
s3, setting coordinates of two pixel points as coordinates of two ends of a reference line through parameter setting by the image recognition anti-collision algorithm system in a picture shot by a camera, calculating the outline of the vehicle to be washed by using a PointRend algorithm, and comparing whether the outline of the vehicle to be washed is crossed with the reference line;
and S4, the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 and informs the user.
The code scanning wake-up algorithm in step S1 includes the following steps:
when a user scans a code, the server acquires a code scanning action of the user, reads the PLC state of the car washer through the Internet of things module, controls the camera to capture an image if the PLC acquires a parking signal of the correlation photoelectric sensor, and wakes up an algorithm to start calculation; if the PLC does not obtain the parking signal of the opposite-emitting photoelectric sensor, the user is prompted through an LED display screen or a mobile phone, the vehicle to be washed is parked in the specified front and rear limited parking areas until the PLC obtains the positions of the front and rear parking areas, and then the image recognition anti-collision algorithm system is awakened. The code that the user scanned is the two-dimensional code that is used for starting the car washer.
The step S1 of reading the car washer PLC state awakening algorithm includes the following steps:
the correlation photoelectric sensor is used for detecting whether the vehicle to be washed enters a front and rear limited parking area, and if a signal fed back to the PLC by the correlation photoelectric sensor indicates that the front and rear limited parking area is indicated, the front and rear position of the vehicle belongs to a safe area; and the upper computer reads the stop signal of the PLC through the Internet of things module and awakens the image recognition anti-collision algorithm system.
The steps can indicate that the front position and the rear position of the vehicle belong to a safe area, but whether the left position and the right position meet the safe standard or not, an image recognition anti-collision algorithm system is needed to realize the steps, and therefore after the signals of the front parking condition and the rear parking condition are obtained, the algorithm is wakened to carry out the next operation.
Further, as shown in fig. 2, the step S3 further includes the following steps:
a straight line is drawn between the two pixel point coordinates set through the parameter setting to serve as a reference line 2;
if any pixel point is superposed with any coordinate point on the reference line in the outline 1 of the vehicle to be washed drawn by the PointRend algorithm, the vehicle to be washed is considered to be out of range; and if all contour pixel points of the vehicle to be washed are not superposed with any coordinate point on the reference line, the vehicle is considered not to be out of range.
Pixel coordinates are a computer-generic definition: the picture in the computer is composed of pixel (the pixel is the general definition of the computer), each pixel has a coordinate position in each picture, the picture is composed of pixels with different colors and different coordinate points and is seen by naked eyes, and each pixel counts coordinates from the upper left corner. A straight line may be defined by defining two points in the image, which are the pixel coordinates defined in step S3 described above. And drawing a straight line between the coordinates of the two pixel points to define the reference line in the invention.
Wherein fig. 2(a) shows a real-time image taken from the rear left of the vehicle to be washed, and fig. 2(b) shows a real-time image taken from the rear right of the vehicle to be washed.
In addition, the step S4 further includes the steps of:
the user wakes up the algorithm by scanning the code, and the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 to the mobile phone end of the user and prompts whether the vehicle to be washed is out of bounds; and (4) the user wakes up the algorithm by reading the state of the car washer PLC, the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 to the PLC, and the PLC outputs a comparison result signal to the LED display screen and informs the user whether the car to be washed is out of range or not.
The system can accurately judge whether the parking position of the vehicle to be washed in the unmanned vehicle washing device is out of range or not, and timely inform a user of vehicle alignment so as to prevent the vehicle from being collided; the invention applies the PointRend algorithm to an unmanned car washing image recognition system, recognizes the car body outline through the algorithm, pre-judges whether the car is out of range or not by presetting whether an anti-collision reference line and the car body outline are crossed or not in the image, and feeds back the result, thereby having the characteristics of high efficiency and high quality.
The foregoing has outlined rather broadly the preferred embodiments and principles of the present invention and it will be appreciated that those skilled in the art may devise variations of the present invention that are within the spirit and scope of the appended claims.
Claims (9)
1. The unmanned car washing image recognition anti-collision system based on the PointRend algorithm is characterized by comprising a car to be washed, a car washer, an image acquisition device, an image recognition anti-collision algorithm system and an information feedback system; the image acquisition devices are arranged at the two sides in front of or behind the car washing station and are used for acquiring photos of the two sides of the car body to be washed; the image acquisition device, the image recognition anti-collision algorithm system and the information feedback system are sequentially in communication connection; the image recognition anti-collision algorithm system comprises a PointRend algorithm module, identifies the vehicle body outline of the vehicle to be washed through the PointRend algorithm, pre-judges whether the vehicle to be washed crosses the boundary or not by presetting whether an anti-collision reference line is crossed with the vehicle body outline or not in the image, and transmits a feedback result to the information feedback system; the information feedback system is used for informing a feedback result to a user.
2. The unmanned vehicle washing image recognition anti-collision system based on the PointRend algorithm is characterized in that the information feedback system comprises at least two pairs of correlation electro-optical sensors, a Programmable Logic Controller (PLC), an Internet of things module, an upper computer, an LED display screen and a server; the server is connected with the Internet of things module, the Internet of things module is connected with the PLC, the PLC is connected with the image acquisition device and each correlation photoelectric sensor respectively, the upper computer is connected with the Internet of things module, and the LED display screen is connected with the PLC.
3. The PointRend algorithm-based unmanned car wash image recognition collision avoidance system of claim 2, wherein the correlation electro-optical sensor and the PLC are both mounted on a car wash.
4. The PointRend algorithm-based unmanned car washing image recognition anti-collision system according to claim 1, 2 or 3, wherein the image acquisition device comprises at least two cameras; the cameras are arranged on the two sides in front of or behind the car washing station; the cameras are in communication connection with the image recognition anti-collision algorithm system.
5. The identification method of the unmanned car washing image identification anti-collision system based on the PointRend algorithm is characterized by comprising the following steps:
s1, driving the vehicle to be washed into the station by the user, and waking up the image recognition anti-collision algorithm system by the user through a code scanning wake-up algorithm or a PLC state read wake-up algorithm of the car washer;
s2, after the algorithm is awakened, the unmanned vehicle washing image recognition anti-collision system collects all pictures shot by the cameras and submits the pictures to an interface of the image recognition anti-collision algorithm system;
s3, setting coordinates of two pixel points as coordinates of two ends of a reference line through parameter setting by the image recognition anti-collision algorithm system in a picture shot by a camera, calculating the outline of the vehicle to be washed by using a PointRend algorithm, and comparing whether the outline of the vehicle to be washed is crossed with the reference line;
and S4, the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 and informs the user.
6. The identification method of the unmanned car washing image identification anti-collision system based on the PointRend algorithm as claimed in claim 5, wherein the code scanning wake-up algorithm in step S1 comprises the following steps:
when a user scans a code, the server acquires a code scanning action of the user, reads the PLC state of the car washer through the Internet of things module, controls the camera to capture an image if the PLC acquires a parking signal of the correlation photoelectric sensor, and wakes up an algorithm to start calculation; if the PLC does not obtain the parking signal of the opposite-emitting photoelectric sensor, the user is prompted through an LED display screen or a mobile phone, the vehicle to be washed is parked in the specified front and rear limited parking areas until the PLC obtains the positions of the front and rear parking areas, and then the image recognition anti-collision algorithm system is awakened.
7. The identification method of the unmanned car washing image identification anti-collision system based on the PointRend algorithm as claimed in claim 5, wherein the step S1 of reading the PLC state awakening algorithm of the car washing machine comprises the following steps:
the correlation photoelectric sensor is used for detecting whether the vehicle to be washed enters a front and rear limited parking area, and if a signal fed back to the PLC by the correlation photoelectric sensor indicates that the front and rear limited parking area is indicated, the front and rear position of the vehicle belongs to a safe area; and the upper computer reads the stop signal of the PLC through the Internet of things module and awakens the image recognition anti-collision algorithm system.
8. The identification method of the unmanned car washing image identification anti-collision system based on the PointRend algorithm as claimed in claim 6 or 7, wherein the step S3 further comprises the steps of:
drawing a straight line between the coordinates of the two pixel points set by the parameter setting to be used as a reference line;
if any pixel point is superposed with any coordinate point on the reference line in the outline of the vehicle to be washed drawn by the PointRend algorithm, the vehicle to be washed is considered to be out of range; and if all contour pixel points of the vehicle to be washed are not superposed with any coordinate point on the reference line, the vehicle is considered not to be out of range.
9. The identification method of the unmanned car washing image identification anti-collision system based on the PointRend algorithm as claimed in claim 8, wherein the step S4 further comprises the following steps:
the user wakes up the algorithm by scanning the code, and the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 to the mobile phone end of the user and prompts whether the vehicle to be washed is out of bounds; and (4) the user wakes up the algorithm by reading the state of the car washer PLC, the image recognition anti-collision algorithm system feeds back the comparison result obtained in the step S3 to the PLC, and the PLC outputs a comparison result signal to the LED display screen and informs the user whether the car to be washed is out of range or not.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011553858.2A CN112507970A (en) | 2020-12-24 | 2020-12-24 | Unmanned car washing image recognition anti-collision system and recognition method based on PointRend algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011553858.2A CN112507970A (en) | 2020-12-24 | 2020-12-24 | Unmanned car washing image recognition anti-collision system and recognition method based on PointRend algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112507970A true CN112507970A (en) | 2021-03-16 |
Family
ID=74923342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011553858.2A Pending CN112507970A (en) | 2020-12-24 | 2020-12-24 | Unmanned car washing image recognition anti-collision system and recognition method based on PointRend algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112507970A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114475530A (en) * | 2021-12-31 | 2022-05-13 | 深圳市小马控股有限公司 | Vehicle washing method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107757563A (en) * | 2017-09-07 | 2018-03-06 | 上海氦修网络科技有限公司 | A kind of car-washing method of automatic car washing machine |
| CN109629873A (en) * | 2018-11-30 | 2019-04-16 | 昆明昆船物流信息产业有限公司 | A kind of automobile vehicle guiding system and method for wisdom parking vehicle access port |
| CN111243024A (en) * | 2020-01-14 | 2020-06-05 | 上海钧正网络科技有限公司 | Parking area identification system and method, detection system and method and security system |
| CN111627253A (en) * | 2020-06-12 | 2020-09-04 | 浙江驿公里智能科技有限公司 | Anti-collision system and method for actively guiding parking by using camera |
-
2020
- 2020-12-24 CN CN202011553858.2A patent/CN112507970A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107757563A (en) * | 2017-09-07 | 2018-03-06 | 上海氦修网络科技有限公司 | A kind of car-washing method of automatic car washing machine |
| CN109629873A (en) * | 2018-11-30 | 2019-04-16 | 昆明昆船物流信息产业有限公司 | A kind of automobile vehicle guiding system and method for wisdom parking vehicle access port |
| CN111243024A (en) * | 2020-01-14 | 2020-06-05 | 上海钧正网络科技有限公司 | Parking area identification system and method, detection system and method and security system |
| CN111627253A (en) * | 2020-06-12 | 2020-09-04 | 浙江驿公里智能科技有限公司 | Anti-collision system and method for actively guiding parking by using camera |
Non-Patent Citations (1)
| Title |
|---|
| ALEXANDER KIRILLOV等: "PointRend: Image Segmentation as Rendering", HTTPS://ARXIV.ORG/PDF/1912.08193.PDF, 16 February 2020 (2020-02-16), pages 2 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114475530A (en) * | 2021-12-31 | 2022-05-13 | 深圳市小马控股有限公司 | Vehicle washing method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107014822B (en) | Non-woven fabric mask defect visual detection system and method | |
| CN111476169B (en) | Complex scene road side parking behavior identification method based on video frame | |
| CN104933409B (en) | A kind of parking stall recognition methods based on panoramic picture dotted line feature | |
| CN108765970B (en) | Method and device for automatically identifying license plate of automobile during roadside parking | |
| CN110217205A (en) | Vehicle cleaning method and system based on image recognition | |
| JP6222948B2 (en) | Feature point extraction device | |
| CN104700114A (en) | Method and device for recognizing pedestrian and vehicle supporting the same | |
| CN102393906B (en) | Computer vision inspection technology-based method for rapid detection and extraction of two-dimensional code | |
| CN105718031A (en) | Gesture recognition method and device | |
| CN104637342A (en) | Intelligent identification and parking path planning system and method for narrow and vertical parking space scene | |
| US20160180158A1 (en) | Vehicle vision system with pedestrian detection | |
| CN105564379A (en) | Automatic starting system and method for automobile windshield wiper | |
| CN110334625A (en) | A kind of parking stall visual identifying system and its recognition methods towards automatic parking | |
| CN112507970A (en) | Unmanned car washing image recognition anti-collision system and recognition method based on PointRend algorithm | |
| CN111881832A (en) | Lane target detection method, device, equipment and computer readable storage medium | |
| CN207718641U (en) | A kind of intelligent machine garage access vehicle monitoring system | |
| CN111968132A (en) | Panoramic vision-based relative pose calculation method for wireless charging alignment | |
| CN113808087A (en) | Defect management and control method and device for surface of steel plate and computer readable storage medium | |
| CN111739333B (en) | Empty parking space identification method | |
| CN111626225A (en) | License plate recognition method, device and equipment for station vehicle and storage medium | |
| CN111274872B (en) | Video monitoring dynamic irregular multi-supervision area discrimination method based on template matching | |
| CN206133727U (en) | District garage intelligent entrance guard management system based on license plate detects | |
| CN115256059B (en) | Grinding disc device control method and system and grinding polisher | |
| CN111369806A (en) | Method and device for photographing, measuring speed and identifying license plate of freight train | |
| CN111152762A (en) | Intelligent car washing control system for construction site |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |